QUALIFICATION MISMATCH AND OCCUPATIONAL CHANGE: THE ROLE OF DEMAND, SUPPLY AND INSTITUTIONAL FACTORS IN EXPLAINING CROSS-NATIONAL VARIATIONS

The occupational structure of advanced economies has undergone significant changes over the past decades. On the demand side, technological change intensified and has been considered the main driver behind these transformations, with competing predictions concerning its impact on society. On the supply side, a very substantial educational expansion took place worldwide, allowing to cope with technological progress, boosting productivity and growth. The interaction between the two creates reasons for optimism through the upgrading of occupational structure, and pessimism due to the “hollowing out” of susceptible to automation middle-skilled jobs and the growth in low-skilled jobs. Furthermore, the imperfect match between demand and supply resulted in an increasing phenomenon of qualification mismatch and overeducation, questioning whether having higher education results in having a better job. However, the magnitude of these processes appears to vary across countries, with more promising upgrading trends in some countries and polarization in others. This suggests that a more detailed assessment of the role of demand, supply factors, and institutional characteristics might be important for drawing conclusions and developing policy remedies to address the challenges posed by rapid technological advancements. In this dissertation, I analyse how the occupational structure has changed in 29 European countries over the past 15 years, investigating whether the technological change has been skill-biased or routine-biased and how these changes in occupational structure affected the risk of overeducation for individuals with different levels of education. I do so by analysing different sources of data and primarily the European Labour Force Survey (EU LFS). In the first part of the thesis, I provide a comparative assessment of changes in occupational structure based on different dimensions of job quality and task content of occupations. This assessment is followed by a multivariate regression analysis analyzing whether the cross-country dynamics of occupational change are explained by the different labor market institutions. In the second part of the thesis, I focus on the relationship between occupational structure and overeducation. I analyse how the incidence of overeducation varied across occupations and task-based occupational groups from a cross-country perspective. Finally, using a mixed-effects linear probability model, I explore how occupational change affects the risks of overeducation for different educational groups. The findings suggest that changes in occupational structure can be simultaneously described as skill- and routine-biased, because the latter successfully explains the decline in the middle that the former does not account for. Routinization does not appear to preclude upgrading, and polarization is determined by the position of routine-intense occupations in the country-specific wage structure, explaining cross-country variation of occupational patterns. Stronger upgrading is noticeable in the countries with higher investments in R&D, education, training, and adult training. Stricter employment protection legislation for regular contracts and collective dismissals mitigates the “hollowing out” in the middle, while the impact of the minimum wage on employment in low-skilled jobs appears to be limited. The evidence further suggests that upgrading of occupational structure and upskilling were accompanied by a rising incidence of overeducation, particularly in the period 2003-2010. Although it slightly declined in the period 2011-2018, a higher incidence of overeducation persisted among the tertiary-educated workers, especially in lower-skilled mildly-routine service occupations with a high degree of interactive tasks. Furthermore, while the risk of overeducation slightly declined for medium-skilled, it has not only persisted but increased for tertiary educated. These results point to a substitution effect, supporting a positional value of education for the labor queue: tertiary skilled workers substitute those medium-skilled, and the medium-skilled substitute the lower-skilled in mildly-routine intense occupations. Since these occupations are difficult to automate and are predicted to expand, the problem of overeducation is likely to persist in the next decades.